Text

SEGMENTASI DAN DETEKSI KELAINAN JANTUNG ANAK DENGAN PANDANGAN 4 CHAMBER VIEW

---

Penilaian

---

Recently, Deep Learning-based approaches in the medical field have brought excellent performance in segmentation and object detection. In computer-based methods, image segmentation is a fundamental way to extract useful information from medical images. Object detection, as the core of the computer vision research problem, has attracted much attention from researchers. Research related to infant heart defects has not been commonly conducted because the data of infant heart medical images are very few to be accessed by many people. The data used is a infant heart ultrasonography video with a four chamber view. The data is converted into frames, annotations, and data augmentation before use. Convolutional Neural Network has been partially applied with promising results on various biomedical imaging problems. In medical image segmentation in recent years most of the credit goes to Unet. As for the case of object detection, Faster R-CNN is one of the best ways to detect objects. In the case of segmentation research conducted, the best model gets an IoU of 97.76 and DSC of 97.89. In the case of object detection, the feature extraction used is VGG16 and ResNet50. In the first detection case performed, the model using Resnet50 was better than VGG16. The model obtained a mAP of 95.67%. However, in the second detection case, the model using VGG16 was the best model with an mAP of 82.38%.

---

Availability

Inventory Code	Barcode	Call Number	Location	Status
2207004829	T82555	T825552022	Central Library (Referens)	Available but not for loan - Not for Loan

Detail Information

Series Title

-

Call Number

T825552022

Publisher

Inderalaya : Jurusan Sistem Komputer, Fakultas Ilmu Komputer Universitas Sriwijaya., 2022

Collation

xix, 82 hlm.; ilus.; 29 cm

Language

Indonesia

ISBN/ISSN

-

Classification

006.307

Content Type

-

Media Type

-

Carrier Type

-

Edition

-

Subject(s)

Sistem Pakar
Jurusan Sistem Komputer

Specific Detail Info

-

Statement of Responsibility

SEPTA

Other version/related

File Attachment

Comments

---

You must be logged in to post a comment